Wayne Vore

I started out with a question: how does this season’s more condensed schedule compare to a normal season as far as the number of games played per day? A typical season is 82 games in 169 days (2011 being typical) and this season is 66 games in 123 days.

Then I wondered about Tim’s minutes and how they compared and the result is that even though Tim is playing a career-low 28.1 minutes per game, he is still playing too much.

Put your math hat on, but don’t be intimidated as I’ll go slow and show my work, and come on in.

My Premise

Here’s what I’m thinking. Last year, Tim played a career-low 28.4 minutes per game. Let’s say, for the sake of argument (and to have something, anything to base analysis on), that Tim entered the playoffs in good shape physically. Let’s assume that Pop wants Tim to enter the playoffs this year in the same condition. So, you are Pop sitting around looking at the team’s schedule and you are scheming to keep Tim fresh come playoff time. Like you did last year.

Getting A Baseline

Since this year’s schedule is a little different, you have to find a way to compare last year’s schedule to this year’s schedule (I’m simplifying this right now by considering last year’s ending schedule average, but I’ll un-simplify it later). This year, team’s are playing a more condensed schedule. In 2011, the Spurs played their 82 game season in the 169 days from October 27th to April 13th (5 in Oct, 30 in Nov, 31 in Dec, 31 in Jan, 28 in Feb, 31 in Mar, and 13 in April). This year, 2012, the Spurs are playing 66 games in the 123 days from Dec 26th to April 26th (6 in Dec, 31 in Jan, 29 in Feb, 31 in Mar, and 26 in April).

If we do a little math, we can compare the days/game density of the two seasons:

  • 2011: 169 days/82 games = 2.06 days/game
  • 2012: 123 days/66 games = 1.86 days/game

As you can see, the number of days per game is lower in 2012. Now, that 0.20 difference might not look like much, but it is almost 10%. It means that in 2012, the Spurs will play 10 games in 18 days instead of in 20 (more or less). That’s two less days of rest for 10 games. Over the course of a 66 game schedule, that’s just over 13 less rest days. Yes, 13 less rest days for Tim.

In Search Of A Meaningful Statistic

Now that we have a metric for comparing the two seasons, I need to find a way to apply this to Tim. In a meaningful way. Here’s what I came up with and I think it holds water.

Minutes per game is not really an accurate measure of wear and tear. Why? Because schedules are not created equal. Wear and tear is really a product of time *on* the court balanced with time *off* the court. What we are really interested in, from the stand point of keeping Tim fresh, is his minutes played per *day*. Minutes per day accounts for things like back to backs, back to back to backs, and four games in five nights.

For instance, if the Spurs are in a four games in five nights stretch, and Tim plays 25 minutes in each game, then he has played 100 minutes in five days. Now, let’s say Pop sits Tim one of those games and then plays Tim 35 minutes in the three games he plays. Tim has now played 105 minutes in five days. Even though he got a game off, he’s actually spent an extra 5 minutes on the court. I’d argue that, in this case, Tim has suffered *more* wear and tear than he would have without the game off.

Minutes per game is misleading. One, because the league does *not* count Tim’s game off in his average minutes. Minutes per game is calculated from games played by the player, not by games played by the team. Two, because it doesn’t account for game density.

Instead of looking at minutes per game, I propose we look at minutes per day.

Tim’s Minutes Per Day

Now that I am fully convinced that I have a meaningful metric, and I am, let’s look at Tim’s 2011 season and figure out how much Pop really played Tim, then let’s compare that to how much Pop is playing him this year.

In 2011, Tim played 2158.4 minutes in the season. The Spurs’ 2011 season was played in 169 days. For Tim, that means, he played:

  • 2158.4 minutes/169 days = 12.77 minutes/day

For every calendar day of the 2011 season, Tim’s right knee logged 12 minutes and 45 seconds of game time. How does that compare to this year? Well, let’s look:

Through the Bobcats’ game on Friday night, Tim had played 955.4 minutes in 68 days (Dec 26 through, and including, March 2nd). That makes:

  • 955.4 minutes/68 days = 14.05 minutes/day

There is a statistical vagary here that I’d like to account for. If I do this same calculation, and include today’s day off (making it 69 days by including March 3rd), then the result is:

  • 955.4 minutes/69 days = 13.85 minutes/day

In order to not skew the numbers more in my favor than necessary, let’s include this day off. This way, we are thinking about how many minutes Tim should play on any given game day. And since today isn’t a game day, let’s look at it as if it was tomorrow. Silly? Maybe. But, I believe it is more fair.

Now, as you can see from the numbers above, Tim is playing a full minute (and about five more seconds) per day. That’s quite a lot. In the old minutes per game world, that’s quite a bit. It’s an extra 2.41 minutes per game. The math:

  • 13.85-12.77 = 1.08 more minutes per day.
  • 1.08/12.77 is 8.5% more minutes per day.
  • Tim’s 2011 minutes per game is 28.4, so 8.5% of 28.4 is 2.41.

In short, considering the abnormal season, Tim is actually playing *more* this year than last year. That tell’s me that Pop needs to reign Tim’s minutes back if he wants Tim to be as fresh in the playoffs as he was last year.

But Wait, There’s More

There is a lurking beast on the horizon. That beast is the Spurs’ finishing stretch of schedule. They play 16 games in the last 24 days of the season. A truly brutal stretch. Now, given that we know Pop doesn’t want to overplay Tim, we have to ask ourselves, “How much should Pop play Tim?”

Well, in order to be as fresh for the playoffs as he was in 2011, we have to conclude that Tim should play 12.77 minutes/day. With 24 days, that means:

  • 12.77 minutes/day * 24 days = 306.48 minutes

If Tim were to play all 16 games, it would mean 19.16 minutes/game. This is very little game time for Tim. How many games would Pop have to rest Tim in order for Tim to play 30 minutes per game in the games that he plays? The answer: Six games. The math:

  • 306.48 minutes / 30 minutes = 10 games played
  • 16 games total – 10 games played = 6 games of rest


As you can see, Pop will have to sit Tim a lot — and I mean a lot — to not wear Tim out before the season ends. He’s already playing Tim more than last year and, in my opinion, flirting with wearing him down too much.

Pop’s a very smart and very conservative (when it comes to Tim’s health) man. If he wants ┬áTim to be ready for the playoffs, he’ll need to keep Tim on the bench a whole lot more than we are used to seeing on a per game basis. Now, I believe, we have a metric for evaluating Tim’s wear and tear and for being able to assess what Pop is doing.

Subscribe To Get Push Notifications For New Posts

2 Responses to Pop Is Playing Tim (A Little) Too Much

  1. LasEspuelas says:

    I like what you have done here. A few comments:

    1. Are all minutes created equal? If we are in a b2b and Timmy has played 25 minutes in the first game what do you think is better, a) play him an extra 10 minutes today and rest him the second game, or b) bench him for the rest of the game and play him 25 minutes the next day? According to your metric, it is better to play him just one day in this particular scenario.

    2. How much does Pop would like to play Timmy if wear and tear were not a concern. From last years playoffs, it seems that number is 36 minutes per game or so. Would it be better to a) play Timmy less games but every time he plays give him a full “playoff workload”, or b) play him more games but less minutes (Pop’s current approach)?

    3. If you decide for 2b, wouldnt it be wise to play Timmy say 25 mpg in the playoffs anyways? Did Duncan get winded by the end of the Memphis series? Hard to say (limited data) but his shooting % in the first two games is markedly higher than the rest of the series.

    4. Finally how productive is Duncan on a per minute basis as a functions of minutes played. Is he more productive in the first ten minutes he plays of the game? How does that compare to minutes 25 to 30, and minutes 30 to 35?

  2. Wayne Vore says:

    1. Hard to say. On the one hand, 25 and 25 is a hell of a lot more than 35 and 0. 20 and 20 is a fair bit more than 35 and 0 because of warm ups, etc. But, those extra 10 minutes from 25 to 35 are played on more tired legs. I’d call it a draw.
    To sum up, I don’t have evidence, but I’m going with all minutes equal.
    2. A hypothetical with no useful answer. Tim’s old. You have to be concerned with wear and tear. If not, Pop would play Tim 48. He’s still that good. But, for your question, I’m going with 2b. Tim’s presence makes the team better. His 20 minutes in a game can be enough to swing it in your favor.
    3. In the playoffs they have more rest. You could probably play Tim 35 mpg and still have him at 13 minutes per day. :) See how I did that.
    4. Tim’s production, imo, isn’t related to the minute in the game. He’s productive when the team needs him and they get him the ball. The issue isn’t Tim’s conditioning. In any given game, Tim can play 35-40 minutes well. The issue is the accumulation of minutes on his knee. I’d ask, is Tim as productive in minutes 2100-2200 of the season as he is in minutes 2400-2500. Something like that.